Premium
Is the oral health impact profile measuring up? Investigating the scale’s construct validity using structural equation modelling
Author(s) -
Baker S. R.,
Gibson B.,
Locker D.
Publication year - 2008
Publication title -
community dentistry and oral epidemiology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.061
H-Index - 101
eISSN - 1600-0528
pISSN - 0301-5661
DOI - 10.1111/j.1600-0528.2008.00440.x
Subject(s) - construct (python library) , construct validity , scale (ratio) , structural equation modeling , conceptual model , measure (data warehouse) , medicine , psychometrics , data mining , statistics , mathematics , computer science , clinical psychology , database , physics , quantum mechanics , programming language
– Objectives: The aim of the study was to provide an empirical test of the construct validity of the Oral Health Impact Profile as a measure of Locker’s conceptual model of oral health. Methods: A secondary analysis of data from the Ontario Study of Older Adults was carried out using structural equation modelling to assess the degree to which scale items measured the construct they were supposed to measure ( within‐ construct validity) and whether relations between constructs were as hypothesized by Locker’s model ( between‐ construct validity). Results: The findings indicated that the Oral Health Impact Profile as currently conceived does not have adequate within‐construct validity. Scale items did not always measure the construct they were supposed to measure, some items within a construct were redundant, many measured more than one construct, and the scale did not represent seven separate constructs of oral health as originally devised. Following reconceptualization of the scale, the revised six‐factor 22 item version was a better fit to the data. However, the scale did not have adequate between‐construct validity. Conclusion: The present findings do not provide support for the conceptual basis of the Oral Health Impact Profile as a measure of Locker’s model of oral health. The need for further conceptual development of the scale, and Locker’s model, are discussed.